Parallel coordinate descent methods for big data optimization

Peter Richtárik*, Martin Takáč

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

152 Scopus citations

Abstract

In this work we show that randomized (block) coordinate descent methods can be accelerated by parallelization when applied to the problem of minimizing the sum of a partially separable smooth convex function and a simple separable convex function. The theoretical speedup, as compared to the serial method, and referring to the number of iterations needed to approximately solve the problem with high probability, is a simple expression depending on the number of parallel processors and a natural and easily computable measure of separability of the smooth component of the objective function. In the worst case, when no degree of separability is present, there may be no speedup; in the best case, when the problem is separable, the speedup is equal to the number of processors. Our analysis also works in the mode when the number of blocks being updated at each iteration is random, which allows for modeling situations with busy or unreliable processors. We show that our algorithm is able to solve a LASSO problem involving a matrix with 20 billion nonzeros in 2 h on a large memory node with 24 cores.

Original languageEnglish (US)
Pages (from-to)433-484
Number of pages52
JournalMathematical Programming
Volume156
Issue number1-2
DOIs
StatePublished - Mar 1 2016

Keywords

  • Big data optimization
  • Composite objective
  • Convex optimization
  • Expected separable over-approximation
  • Huge-scale optimization
  • Iteration complexity
  • LASSO
  • Parallel coordinate descent
  • Partial separability

ASJC Scopus subject areas

  • Software
  • Mathematics(all)

Fingerprint Dive into the research topics of 'Parallel coordinate descent methods for big data optimization'. Together they form a unique fingerprint.

Cite this